Work on the central nervous system control of the cardiovascular system has also provided new insights into long standing problems. Most research on both pathophysiological disorders such as hypertension and psychopathological disorders such as anxiety and depression has viewed the cardiovascular and autonomic concomitants of these disorders in terms of sympathetic nervous system activity. However, our work based upon solid theoretical and empirical foundations has sought to re-frame the extant literature in terms of a parasympathetic or vagal deficit (Thayer and Lane, 2000). Individuals with these disorders have been shown to have decreased parasympathetic control of the cardiovascular system. This work has helped to clarify the increased risk of morbidity and mortality associated with these disorders. Recent work has shown how this re-framing may be instantiated in a computational neural network model of cardiac and emotional regulation. Based upon our previous experimental work we have hypothesized a reciprocal inhibitory cortical-subcortical network associated with autonomic, emotional, and cognitive regulation. We modelled this network with Affective Feature units that were analogous to amygdala activity and Decision/Output units that were analogous to activity of the prefrontal cortex. Activity in the Decision units inhibited acivity in the Affective units as an analog to the idea that prefrontal cortex inhibits amygdala activity. Output was modelled as changes in vagally-mediated heart rate variability (HRV). We presented this network with emotionally positive, negative, and neutral inputs and observed the effects on the Affective, Decision, and output units. In response to positive or negative inputs, activity in the Affective Feature units (amygdala) quickly increased and rapidly decayed relative to neutral inputs. Similarly, activity in the Decision units (prefrontal cortex)in response to positive or negative inputs relative to neutral inputs increased after a short delay and then decreased as would be expected in this reciprocal inhibitory network. HRV decreased in response to positive or negative inputs relative to neutral inputs. Importantly when depression was modelled by both over-exposure to negative inputs (as might be the case when people ruminate) and by decreased inhibitory input from the Decision units, activity in the Affective units was increased in response to negative inputs relative to positive or neutral and HRV was concomitantly decreased. Thus this computational network supports our model of neurovisceral integration in which activity in a reciprocal inhibitory cortical-subcortical circuit can be indexed by HRV. Individuals with decreased parasympathetic control of the heart show less ability to adapt both physiologically and psychologically to environmental demands leading to a rigid, inflexible response disposition. Recent studies have extended this model to cognitive processes including working memory. Results indicate that those with higher HRV perform better on tasks that require executive processes and working memory. These findings may guide us as we examine the central and peripheral concomitants of autonomic, emotional, and cognitive regulation.